Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add New-Kringster/shebang-mcp --skill shebanggit clone --depth 1 https://github.com/New-Kringster/shebang-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/new-kringster/shebang-mcp/shebang)<a href="https://agentmods.dev/skills/new-kringster/shebang-mcp/shebang"><img src="https://agentmods.dev/badge/skills/new-kringster/shebang-mcp/shebang/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/new-kringster/shebang-mcp/shebang"><img src="https://agentmods.dev/badge/skills/new-kringster/shebang-mcp/shebang.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00175 | $0.02828 |
| Opus 5 | $0.00088 | $0.01414 |
| Sonnet 5 | $0.00035 | $0.00566 |
| Haiku 4.5 | $0.00017 | $0.00283 |
Grade A, and why
shebang scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.
shebang
shebang is one account, one API key, and one MCP server covering five
things an agent commonly needs while working: hosting pages, sharing
files, minting short links, running a Postgres database, and
managing its own identity (API keys, OAuth clients). All of it is
reachable through the shebang-mcp server's tools — no separate accounts,
no separate SDKs.
Full docs, for anything this file doesn't cover in enough depth, are at
https://docs.shebang.pro — including an agent-oriented index at
https://docs.shebang.pro/llms.txt, and every page there fetchable as
plain Markdown at <page-url>.md instead of rendered HTML.
Login-if-401
Every tool call in this server needs an shb_… API key. If a tool call
fails with something like "missing API key" or 401/invalid_api_key, run:
npx -y shebang-mcp login
This starts a device-login flow: it prints a code and a URL, tries to open
a browser, and polls until a human approves it there. On approval it writes
~/.config/shebang/credentials.json (mode 0600) and every future tool call
in this server (or any other harness pointed at the same machine) picks the
key up automatically — log in once, not per-project. If a browser isn't
available on this machine, open the printed URL from any other device and
approve there; the CLI keeps polling either way.
Projects
A Project groups a key's resources — keys, databases, files, links, and
pages — for dashboard organization and access control; it's unrelated to the
apps/project scope list on a key (page/serve/link/base, see
below). Every key has exactly one home Project, which it always reaches in
full. A master key can name any of the account's other Projects (via a
project argument, a slug); an app key can only reach a Project it's been
explicitly granted (read or full, via project_grant_key) beyond its
own home Project. Naming a Project the key can't reach fails as
insufficient_scope, never a not-found error. Most agents never need to
think about Projects at all — every tool defaults to the key's own home
Project when the optional project argument is omitted. Each call is
scoped independently: passing project on one call doesn't change what a
later call defaults to. Publishing a page with project: "acme" and
then calling link_list with no project still returns your home
Project's links, not acme's — pass project: "acme" again on that call
too.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago Changed · -6 tokens per session d408173aba2a
- 3d ago Changed · +59 lines 3d86c3f10e12
- 4d ago Changed · +19 lines 484993096054
- 10d ago First seen · 132 lines · 181 tokens per session scan A a88384f875fc
shebang is a skill published in the GitHub repository New-Kringster/shebang-mcp (0 stars, last pushed 3d ago), licensed MIT. It adds 175 tokens to every session and 2,828 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
alloydb-basics
Manages clusters, instances, and backups for AlloyDB for PostgreSQL, and integrates with AlloyDB Model Context Protocol (MCP) tools for automated database operations. Use when creating, configuring, or administering AlloyDB databases. Do NOT use for general PostgreSQL instances (e.g. Cloud SQL) or other GCP databases.
postgresql-table-design
Use this skill when designing or reviewing a PostgreSQL-specific schema. Covers best-practices, data types, indexing, constraints, performance patterns, and advanced features.
db-repair
Auto-fix gbrain's Postgres access so the brain stays available. When any gbrain command or MCP tool result carries a GBRAINDBACCESS marker (or an operator reports the brain database is down), run the hardcoded gbrain db-repair ladder: diagnose, apply the safe tier, verify. The action is ALWAYS the hardcoded command …
dsql
Build with Aurora DSQL — manage schemas, execute queries, handle migrations, diagnose query plans, diagnose cluster performance, load data, and develop applications with a serverless, distributed SQL database. Covers IAM auth, multi-tenant patterns, MySQL-to-DSQL and PostgreSQL-to-DSQL schema conversion, foreign key…
analyzing-insights-across-teams
Analyze PostHog insights, dashboards, or teams beyond the current project by querying the prod Postgres replicas synced into the dogfood data warehouse (US project 2, "PostHog App + Website"). Use when asked to analyze insights across all teams or projects, another team's insights, or fleet-wide insight/dashboard…
volcengine-rds-postgresql
A tool for operating PostgreSQL databases hosted by Volcano Engine's managed database service. PostgreSQL is a relational database used to store structured application data.